The transcriptional activation of eukaryotic genes is a highly regulated process, where a large number of proteins must act in concert to ensure the correct spatio-temporal expression of genes. Transcription factors are key players in this control, as they represent the ultimate regulators of signal transduction cascades. They are subject to a strong regulatory control by cells, since they give to the cells of particular tissues a specific proliferative or differentiated state, and allow that they may adapt and react to a modified environment or stimulus.
The response of a cell to a given condition often results in the stimulation of many regulatory networks, and hence the activation or repression of multiple transcription factors. It is generally admitted that it is the co-ordination of all resulting signals rather than their individual effect that leads to the highly specific pattern of gene expression dictating the cellular response to the initial stimulus. Deciphering the components of these regulatory networks is of particular interest if one aims to better understand, and hence be able to manipulate the expression of the genome.
It is however not an easy task, as some transduction cascades can parallel each other, be interconnected, or exhibit compensatory effects.
There is therefore a need for methods allowing to get a global picture of a cellular activation profile in different stress conditions, through the simultaneous monitoring of a high number of intracellular signals. Such methods would produce large datasets of cellular activation profiles, which might be particularly useful in the drug discovery and development programs. They would indeed help predicting the global effects of yet uncharacterised test molecules, highlight early in the development process unexpected side effects, and enable the identification of the main cellular targets. Comparing the activation profiles of cells following different drug treatments could help classifying those drugs, and identify cellular components whose manipulation might potentate drug effects.
A classification of transcription factors based on their function within cellular regulatory circuits has been recently established by Brivanlou and Darnell (Brivanlou A. H. and Darnell J. E. Science 295, 813-818, 2002). As shown in table 1, transcription factors are divided into constitutively active and regulatory factors, the former being present in the cell nucleus of all cells while the latter are produced in a cell- and time-specific manner. The regulatory factors are composed of developmental and signal-dependent transcription factors, the latter requiring an appropriate stimuli to become transcriptionally active. These signal-dependent transcription factors are divided into steroid receptors, factors activated by internal signals, and factors activated by cell surface receptor-ligand interactions. This last family is composed of factors present in the nucleus or in the cytoplasm, which become transcriptionally active upon stimulation. This is accompanied by a cytoplasm-to-nucleus translocation for the factors retained in the cytoplasm in a latent, inactive form. This ‘functional’ classification therefore reflects the connection between extracellular signals and the regulation of transcription in eukaryotic cells. It differs from the classical ‘structural’ classification, where transcription factors are grouped based on the structure of their DNA binding motif.
TABLE 1classification of transcription factors on a functional basisaccording to Brivanlou and Darnell. Families and examples oftranscription factors belonging to each family are presented.The list of transcription factors is not exhaustive, and ispresented for illustration purposes only.regulatorysignal-dependentexternal signals:receptor-ligand interactionsconstitu-develop-steroidinternalresidentlatenttivementalreceptorssignalsnuclearcytoplasmicSp1GATAsGRSREBPEts/Elk-1STATsCCAATHNFsERp53CREBSMADsNF1Pit1PRATF-2NFκBMyoDTRATMsRelMyf5RARsSRFCI/GLIBicoldRXRsFOSNOTCHHoxPPARsJUNTUBBYForkheadMEF2NFATsCbfa1DBP
Transcription factors have been implicated in several human diseases, since they control the expression of genes (some being “protective” or “defensive” and others being responsible for secondary deleterious effects). The regulation of the transcription factors may vary with age or may be altered in pathological situations explaining their effects in diseases such as chronic diseases. Furthermore, the dysfunction of a single factor may modify the expression of several genes and affect dozens of different cell types. Molecular characterization of these altered transcription factors, especially regarding their activity or their interaction with other members of the activation pathway and with the transcriptional initiation complex is therefore of crucial importance to understand their role in a defined pathology and associate them with the pathology. Establishing cell activation patterns based on transcription factor profiling is therefore also of interest to follow disease progression, evaluate the effects of potential treatments, and help identifying the regulatory pathways disrupted.
Establishing the activation profile of a cell through the monitoring of its transcription factor components relies on our ability to discriminate between those factors present in the cell under a resting state, and those rendered transcriptionally active as a result of transduction cascade stimulation. Indeed, the activation of specific transcription factors is largely governed at the post-transcriptional level, and usually depends on post-translational modifications of the factor itself, on its interaction with enzymes or regulatory proteins, or on modifications of its sub cellular localization, the final result being either a stimulation or a repression of one or a series of particular genes.
The most common post-translational modification of transcription factors is phosphorylation/dephosphorylation by kinases/phosphatases, which modulates the DNA-binding capacity, nuclear translocation and/or transcriptional potential of said factors. Protein kinases and phosphatases have been co-purified with transcription factors, suggesting physical association between these proteins. Other modifications have been found of importance for transcription factors activity, such as acetylation, oxidation/reduction, nitrosylation and glycosylation.
Many transcription factors ensure the activation of particular genes through interactions with specific cofactors. As an example, the myogenic basic helix-loop-helix (bHLH) transcription factors direct the expression of their target genes in association with E-proteins such as E12 and MEF2 family members. These protein-protein interactions enhance the activation potential of bHLH transcription factors, and in the case of MyoD and MEF2 cooperativity, the DNA binding of only one of the two partners is required.
Another level of protein-protein interactions mediated by transcription factors connect them to proteins from the basal transcription initiation complex. These can be direct or indirect interactions through cofactors which then interact with the basal complex. This is exemplified by CREB-binding protein (CBP) and the related protein p300, that were found to interact with a growing list of transcription factors. This results in the association of CBP/p300 with components of the basal transcriptional machinery and in the acetylation of histones. It was shown that the intrinsic acetyltransferase activity of these cofactors is also responsible for acetylating the transcription factor itself. CBP/p300, which do not bind DNA by themselves, have therefore emerged as general co-activators, which are recruited by activated transcription factors to the promoter of target genes, where they form a bridge with the basal transcriptional machinery, acetylate histone proteins and transcription factors, and lead to transcriptional activation.
It thus becomes more and more clear that post-translational modifications and protein-protein interactions represent key events in the activation of transcription factors and the control of gene expression. There is therefore intense interest in being able to identify the binding partners of transcription factors and assay their activity. Other proteins exert an important regulation on the transcriptional activation through their DNA or protein interactions, such as histone acetylases, deacetylases, methylases, and chromatin remodelers. It is therefore also of interest to develop methods aimed at detecting and quantifying factors, enzymes and compounds controlling transcription. Such methods should help in the discovery of drugs that interfere with the transcriptional regulation of genes, which can be deregulated in many disorders, and improve the diagnostic of such pathologies through the detection of modifications in the activated state of transcription factors and/or in some of their interactions with protein partners.
The simultaneous assessment of transcription factors present in a transcriptionally active or inactive state, and the building of cellular activation profiles are at the basis of the present invention. The method is amendable to an automated, high throughput format, allowing to process multiple samples.